arXiv subjects

Wei Zhao

Publications and source records attributed to Wei Zhao.

At least 109 records · Page 6Linked to original sources

arXiv2025

Frequency Bias Matters: Diving into Robust and Generalized Deep Image Forgery Detection

As deep image forgery powered by AI generative models, such as GANs, continues to challenge today's digital world, detecting AI-generated forgeries has become a vital security topic. Generalizability and robustness are two critical concerns of a forgery detector, determining its reliability when facing unknown GANs and noisy samples in an open world. Although many studies focus on improving these two properties, the root causes of these problems have not been fully explored, and it is unclear if there is a connection between them. Moreover, despite recent achievements in addressing these issues from image forensic or anti-forensic aspects, a universal method that can contribute to both sides simultaneously remains practically significant yet unavailable. In this paper, we provide a fundamental explanation of these problems from a frequency perspective. Our analysis reveals that the frequency bias of a DNN forgery detector is a possible cause of generalization and robustness issues. Based on this finding, we propose a two-step frequency alignment method to remove the frequency discrepancy between real and fake images, offering double-sided benefits: it can serve as a strong black-box attack against forgery detectors in the anti-forensic context or, conversely, as a universal defense to improve detector reliability in the forensic context. We also develop corresponding attack and defense implementations and demonstrate their effectiveness, as well as the effect of the frequency alignment method, in various experimental settings involving twelve detectors, eight forgery models, and five metrics.

cs.CR↗

arXiv2025

Jet reorientation revealed by intermittent jet activity in radio galaxy 0954+556

Intermittent jet activity of AGNs is a common phenomenon, whereas significant jet reorientation during episodic jet activity in relatively young radio galaxies are rarely reported. The quasar 0954+556 at redshift of 0.903 is an intriguing source exhibiting an unusual radio jet structure with significantly different jet directions at kpc and pc scales. At kpc scales, images from the VLA exhibit a bright core, a linear jet extending 24 kpc to the northwest, and a discrete jet component 16 kpc to the northeast. At pc scales, images from the VLBA show a two-component structure with a projected separation of 360 pc in the north-south direction. The peculiar structure of 0954+556 might result from jet reorientation. Here, our aim was to investigate the possible mechanism via multiscale and multifrequency deep radio images. We performed VLA and VLBA observations of 0954+556. Together with some existing data in the NRAO data archive, we made multiple VLA images at 1.4-22 GHz and VLBA images at 1.7-43 GHz for various image analyses of the jet structure. We identified the location of the radio core at pc scales, detected the faint counter-jets at both pc and kpc scales for the first time, and revealed a diffuse emission region connecting pc and kpc scale forward jets. Our spectral index distribution and spectral aging analysis indicate that 0954+556 might undergo at least two episodes of jet activity during the current AGN phase. Moreover, pc scale polarization maps display a well-resolved spine-sheath polarization structure. It seems that the jet direction of 0954+556 changed significantly during intermittent jet activity. This may explain the different jet orientations and spectral ages observed from kpc to pc scales. The research provides a strong case that AGN jet direction might change rapidly on timescales of one million years.

astro-ph.HE↗

arXiv2025

Hilbert's fourth problem in the constant curvature setting

Hilbert's fourth problem seeks the classification of metric geometries where straight lines are shortest paths. Its regular case identifies the projectively flat Finsler manifolds. This broader framework breaks the equivalence between projective flatness and constant curvature that holds in the Riemannian setting, creating a more intricate classification problem. This paper resolves the long-standing question of how the local structure determines the global topology for such manifolds of constant flag curvature, where flag curvature is the natural generalization of Riemannian sectional curvature. We derive explicit distance formulas for all cases of constant flag curvature. For non-positive constant curvature, we establish a global classification of forward complete manifolds, a uniqueness theorem for forward complete metrics, and a characterization of maximal domains of metrics where exotic examples are constructed. For positive constant curvature, we prove a maximum diameter theorem and show the completion of such manifold is a sphere. A fundamental connection is revealed between Sobolev space nonlinearity and backward incompleteness. This work provides a complete characterization of the global geometry for the regular case of Hilbert's fourth problem with constant flag curvature.

math.DG↗

arXiv2025

Q-MLLM: Vector Quantization for Robust Multimodal Large Language Model Security

Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities in cross-modal understanding, but remain vulnerable to adversarial attacks through visual inputs despite robust textual safety mechanisms. These vulnerabilities arise from two core weaknesses: the continuous nature of visual representations, which allows for gradient-based attacks, and the inadequate transfer of text-based safety mechanisms to visual content. We introduce Q-MLLM, a novel architecture that integrates two-level vector quantization to create a discrete bottleneck against adversarial attacks while preserving multimodal reasoning capabilities. By discretizing visual representations at both pixel-patch and semantic levels, Q-MLLM blocks attack pathways and bridges the cross-modal safety alignment gap. Our two-stage training methodology ensures robust learning while maintaining model utility. Experiments demonstrate that Q-MLLM achieves significantly better defense success rate against both jailbreak attacks and toxic image attacks than existing approaches. Notably, Q-MLLM achieves perfect defense success rate (100\%) against jailbreak attacks except in one arguable case, while maintaining competitive performance on multiple utility benchmarks with minimal inference overhead. This work establishes vector quantization as an effective defense mechanism for secure multimodal AI systems without requiring expensive safety-specific fine-tuning or detection overhead. Code is available at https://github.com/Amadeuszhao/QMLLM.

cs.CR↗

arXiv2025

AutoBackdoor: Automating Backdoor Attacks via LLM Agents

Backdoor attacks pose a serious threat to the secure deployment of large language models (LLMs), enabling adversaries to implant hidden behaviors triggered by specific inputs. However, existing methods often rely on manually crafted triggers and static data pipelines, which are rigid, labor-intensive, and inadequate for systematically evaluating modern defense robustness. As AI agents become increasingly capable, there is a growing need for more rigorous, diverse, and scalable \textit{red-teaming frameworks} that can realistically simulate backdoor threats and assess model resilience under adversarial conditions. In this work, we introduce \textsc{AutoBackdoor}, a general framework for automating backdoor injection, encompassing trigger generation, poisoned data construction, and model fine-tuning via an autonomous agent-driven pipeline. Unlike prior approaches, AutoBackdoor uses a powerful language model agent to generate semantically coherent, context-aware trigger phrases, enabling scalable poisoning across arbitrary topics with minimal human effort. We evaluate AutoBackdoor under three realistic threat scenarios, including \textit{Bias Recommendation}, \textit{Hallucination Injection}, and \textit{Peer Review Manipulation}, to simulate a broad range of attacks. Experiments on both open-source and commercial models, including LLaMA-3, Mistral, Qwen, and GPT-4o, demonstrate that our method achieves over 90\% attack success with only a small number of poisoned samples. More importantly, we find that existing defenses often fail to mitigate these attacks, underscoring the need for more rigorous and adaptive evaluation techniques against agent-driven threats as explored in this work. All code, datasets, and experimental configurations will be merged into our primary repository at https://github.com/bboylyg/BackdoorLLM.

cs.CR↗

arXiv2025

Failure of famous functional inequalities on Finsler manifolds: the influence of $S$-curvature

The validity of functional inequalities on Finsler metric measure manifolds is based on three non-Riemannian quantities, namely, the reversibility, flag curvature and $S$-curvature induced by the measure. Under mild assumptions on the reversibility and flag curvature, it turned out that famous functional inequalities -- as Hardy inequality, Heisenberg--Pauli--Weyl uncertainty principle and Caffarelli--Kohn--Nirenberg inequality -- usually hold on forward complete Finsler manifolds with non-positive $S$-curvature, cf. Huang, Kristály and Zhao [Trans. Amer. Math. Soc., 2020]. In this paper however we prove that -- under similar assumptions on the reversibility and flag curvature as before -- the aforementioned functional inequalities fail whenever the $S$-curvature is positive. Accordingly, our results clearly reveal the deep dependence of functional inequalities on the $S$-curvature. As a consequence of these results, we establish analytic aspects of Finsler manifolds, e.g., if the flag curvature is non-positive, the Ricci curvature is bounded from below and the $S$-curvature is positive, then the reversibility turns out to be infinite. Further topological properties and examples are presented on general Funk metric spaces, where the $S$-curvature plays again a decisive role.

math.DG↗

arXiv2025

Privacy Structure and Blackwell Frontier

This paper characterizes the set of feasible posterior distributions subject to graph-based inferential privacy constraint, including both differential and inferential privacy. This characterization can be done through enumerating all extreme points of the feasible posterior set. A connection between extreme posteriors and strongly connected semi-chains is then established. All these semi-chains can be constructed through successive unfolding operations on semi-chains with two partitions, which can be constructed through classical spanning tree algorithm. A sharper characterization of semi-chains with two partitions for differential privacy is provided.

econ.TH↗

arXiv2025

Knowledge-Augmented Long-CoT Generation for Complex Biomolecular Reasoning

Understanding complex biomolecular mechanisms requires multi-step reasoning across molecular interactions, signaling cascades, and metabolic pathways. While large language models(LLMs) show promise in such tasks, their application to biomolecular problems is hindered by logical inconsistencies and the lack of grounding in domain knowledge. Existing approaches often exacerbate these issues: reasoning steps may deviate from biological facts or fail to capture long mechanistic dependencies. To address these challenges, we propose a Knowledge-Augmented Long-CoT Reasoning framework that integrates LLMs with knowledge graph-based multi-hop reasoning chains. The framework constructs mechanistic chains via guided multi-hop traversal and pruning on the knowledge graph; these chains are then incorporated into supervised fine-tuning to improve factual grounding and further refined with reinforcement learning to enhance reasoning reliability and consistency. Furthermore, to overcome the shortcomings of existing benchmarks, which are often restricted in scale and scope and lack annotations for deep reasoning chains, we introduce PrimeKGQA, a comprehensive benchmark for biomolecular question answering. Experimental results on both PrimeKGQA and existing datasets demonstrate that although larger closed-source models still perform well on relatively simple tasks, our method demonstrates clear advantages as reasoning depth increases, achieving state-of-the-art performance on multi-hop tasks that demand traversal of structured biological knowledge. These findings highlight the effectiveness of combining structured knowledge with advanced reasoning strategies for reliable and interpretable biomolecular reasoning.

cs.AI↗

arXiv2025

Fairness-Aware Multi-view Evidential Learning with Adaptive Prior

Multi-view evidential learning aims to integrate information from multiple views to improve prediction performance and provide trustworthy uncertainty esitimation. Most previous methods assume that view-specific evidence learning is naturally reliable. However, in practice, the evidence learning process tends to be biased. Through empirical analysis on real-world data, we reveal that samples tend to be assigned more evidence to support data-rich classes, thereby leading to unreliable uncertainty estimation in predictions. This motivates us to delve into a new Biased Evidential Multi-view Learning (BEML) problem. To this end, we propose Fairness-Aware Multi-view Evidential Learning (FAML). FAML first introduces an adaptive prior based on training trajectory, which acts as a regularization strategy to flexibly calibrate the biased evidence learning process. Furthermore, we explicitly incorporate a fairness constraint based on class-wise evidence variance to promote balanced evidence allocation. In the multi-view fusion stage, we propose an opinion alignment mechanism to mitigate view-specific bias across views, thereby encouraging the integration of consistent and mutually supportive evidence.Theoretical analysis shows that FAML enhances fairness in the evidence learning process. Extensive experiments on five real-world multi-view datasets demonstrate that FAML achieves more balanced evidence allocation and improves both prediction performance and the reliability of uncertainty estimation compared to state-of-the-art methods.

cs.LG↗

arXiv2025

H-NeiFi: Non-Invasive and Consensus-Efficient Multi-Agent Opinion Guidance

The openness of social media enables the free exchange of opinions, but it also presents challenges in guiding opinion evolution towards global consensus. Existing methods often directly modify user views or enforce cross-group connections. These intrusive interventions undermine user autonomy, provoke psychological resistance, and reduce the efficiency of global consensus. Additionally, due to the lack of a long-term perspective, promoting local consensus often exacerbates divisions at the macro level. To address these issues, we propose the hierarchical, non-intrusive opinion guidance framework, H-NeiFi. It first establishes a two-layer dynamic model based on social roles, considering the behavioral characteristics of both experts and non-experts. Additionally, we introduce a non-intrusive neighbor filtering method that adaptively controls user communication channels. Using multi-agent reinforcement learning (MARL), we optimize information propagation paths through a long-term reward function, avoiding direct interference with user interactions. Experiments show that H-NeiFi increases consensus speed by 22.0% to 30.7% and maintains global convergence even in the absence of experts. This approach enables natural and efficient consensus guidance by protecting user interaction autonomy, offering a new paradigm for social network governance.

cs.SI↗

arXiv2025

ECVL-ROUTER: Scenario-Aware Routing for Vision-Language Models

Vision-Language Models (VLMs) excel in diverse multimodal tasks. However, user requirements vary across scenarios, which can be categorized into fast response, high-quality output, and low energy consumption. Relying solely on large models deployed in the cloud for all queries often leads to high latency and energy cost, while small models deployed on edge devices are capable of handling simpler tasks with low latency and energy cost. To fully leverage the strengths of both large and small models, we propose ECVL-ROUTER, the first scenario-aware routing framework for VLMs. Our approach introduces a new routing strategy and evaluation metrics that dynamically select the appropriate model for each query based on user requirements, maximizing overall utility. We also construct a multimodal response-quality dataset tailored for router training and validate the approach through extensive experiments. Results show that our approach successfully routes over 80\% of queries to the small model while incurring less than 10\% drop in problem solving probability.

cs.LG↗

arXiv2025

HF-VTON: High-Fidelity Virtual Try-On via Consistent Geometric and Semantic Alignment

Virtual try-on technology has become increasingly important in the fashion and retail industries, enabling the generation of high-fidelity garment images that adapt seamlessly to target human models. While existing methods have achieved notable progress, they still face significant challenges in maintaining consistency across different poses. Specifically, geometric distortions lead to a lack of spatial consistency, mismatches in garment structure and texture across poses result in semantic inconsistency, and the loss or distortion of fine-grained details diminishes visual fidelity. To address these challenges, we propose HF-VTON, a novel framework that ensures high-fidelity virtual try-on performance across diverse poses. HF-VTON consists of three key modules: (1) the Appearance-Preserving Warp Alignment Module (APWAM), which aligns garments to human poses, addressing geometric deformations and ensuring spatial consistency; (2) the Semantic Representation and Comprehension Module (SRCM), which captures fine-grained garment attributes and multi-pose data to enhance semantic representation, maintaining structural, textural, and pattern consistency; and (3) the Multimodal Prior-Guided Appearance Generation Module (MPAGM), which integrates multimodal features and prior knowledge from pre-trained models to optimize appearance generation, ensuring both semantic and geometric consistency. Additionally, to overcome data limitations in existing benchmarks, we introduce the SAMP-VTONS dataset, featuring multi-pose pairs and rich textual annotations for a more comprehensive evaluation. Experimental results demonstrate that HF-VTON outperforms state-of-the-art methods on both VITON-HD and SAMP-VTONS, excelling in visual fidelity, semantic consistency, and detail preservation.

cs.CV↗

arXiv2025

An Adaptive Edge-Guided Dual-Network Framework for Fast QR Code Motion Deblurring

Unlike general image deblurring that prioritizes perceptual quality, QR code deblurring focuses on ensuring successful decoding. QR codes are characterized by highly structured patterns with sharp edges, a robust prior for restoration. Yet existing deep learning methods rarely exploit these priors explicitly. To address this gap, we propose the Edge-Guided Attention Block (EGAB), which embeds explicit edge priors into a Transformer architecture. Based on EGAB, we develop Edge-Guided Restormer (EG-Restormer), an effective network that significantly boosts the decoding rate of severely blurred QR codes. For mildly blurred inputs, we design the Lightweight and Efficient Network (LENet) for fast deblurring. We further integrate these two networks into an Adaptive Dual-network (ADNet), which dynamically selects the suitable network based on input blur severity, making it ideal for resource-constrained mobile devices. Extensive experiments show that our EG-Restormer and ADNet achieve state-of-the-art performance with a competitive speed. Project page: https://github.com/leejianping/ADNet

cs.CV↗

arXiv2025

LiTransProQA: an LLM-based Literary Translation evaluation metric with Professional Question Answering

The impact of Large Language Models (LLMs) has extended into literary domains. However, existing evaluation metrics for literature prioritize mechanical accuracy over artistic expression and tend to overrate machine translation as being superior to human translation from experienced professionals. In the long run, this bias could result in an irreversible decline in translation quality and cultural authenticity. In response to the urgent need for a specialized literary evaluation metric, we introduce LITRANSPROQA, a novel, reference-free, LLM-based question-answering framework designed for literary translation evaluation. LITRANSPROQA integrates humans in the loop to incorporate insights from professional literary translators and researchers, focusing on critical elements in literary quality assessment such as literary devices, cultural understanding, and authorial voice. Our extensive evaluation shows that while literary-finetuned XCOMET-XL yields marginal gains, LITRANSPROQA substantially outperforms current metrics, achieving up to 0.07 gain in correlation and surpassing the best state-of-the-art metrics by over 15 points in adequacy assessments. Incorporating professional translator insights as weights further improves performance, highlighting the value of translator inputs. Notably, LITRANSPROQA reaches an adequacy performance comparable to trained linguistic student evaluators, though it still falls behind experienced professional translators. LITRANSPROQA shows broad applicability to open-source models like LLaMA3.3-70b and Qwen2.5-32b, indicating its potential as an accessible and training-free tool for evaluating literary translations that require local processing due to copyright or ethical considerations.

cs.CL↗

arXiv2025

Dynamic Network-Based Two-Stage Time Series Forecasting for Affiliate Marketing

In recent years, affiliate marketing has emerged as a revenue-sharing strategy where merchants collaborate with promoters to promote their products. It not only increases product exposure but also allows promoters to earn a commission. This paper addresses the pivotal yet under-explored challenge in affiliate marketing: accurately assessing and predicting the contributions of promoters in product promotion. We design a novel metric for evaluating the indirect contributions of the promoter, called propagation scale. Unfortunately, existing time series forecasting techniques fail to deliver accurate predictions due to the propagation scale being influenced by multiple factors and the inherent complexities arising from dynamic scenarios. To address this issue, we decouple the network structure from the node signals and propose a two-stage solution: initially, the basic self-sales and network structure prediction are conducted separately, followed by the synthesis of the propagation scale. Specifically, we design a graph convolution encoding scheme based on descendant neighbors and incorporate hypergraph convolution to efficiently capture complex promotional dynamics. Additionally, three auxiliary tasks are employed: self-sales prediction for base estimations, descendant prediction to synthesize propagation scale, and promoter activation prediction to mitigate high volatility issues. Extensive offline experiments on large-scale industrial datasets validate the superiority of our method. We further deploy our model on Alimama platform with over $100,000$ promoters, achieving a $9.29\%$ improvement in GMV and a $5.89\%$ increase in sales volume.

cs.IR↗

arXiv2025

Convergence Guarantees for Gradient-Based Training of Neural PDE Solvers: From Linear to Nonlinear PDEs

We present a unified convergence theory for gradient-based training of neural network methods for partial differential equations (PDEs), covering both physics-informed neural networks (PINNs) and the Deep Ritz method. For linear PDEs, we extend the neural tangent kernel (NTK) framework for PINNs to establish global convergence guarantees for a broad class of linear operators. For nonlinear PDEs, we prove convergence to critical points via the Łojasiewicz inequality under the random feature model, eliminating the need for strong over-parameterization and encompassing both gradient flow and implicit gradient descent dynamics. Our results further reveal that the random feature model exhibits an implicit regularization effect, preventing parameter divergence to infinity. Theoretical findings are corroborated by numerical experiments, providing new insights into the training dynamics and robustness of neural network PDE solvers.

math.NA↗

arXiv2025

An Approach for Restoring Magnetic Field Uniformity in Openable BIPM-Type Kibble Balance Magnets

The Kibble balance realizes the kilogram by linking mechanical and electrical quantities via a magnet system. In an improved BIPM-type magnet design by Tsinghua University, an open/close surface was incorporated, facilitating operation. However, an unavoidable mechanical air gap at the splitting plane introduces asymmetry in the magnetic flux density profile, degrading field uniformity. This study proposes a two-step yoke compensation method to restore symmetry by adjusting the upper outer yoke's inner radius and the splitting gap height. Finite element simulations show linear relationships between asymmetry and these parameters, enabling predictive compensation. Experimental results confirm that sequential tuning successfully eliminates asymmetry and recovers the designed uniform field range. The method provides an effective solution for enhancing magnetic field quality in openable Kibble balance magnets.

physics.ins-det↗

arXiv2025

Where Did It Go Wrong? Attributing Undesirable LLM Behaviors via Representation Gradient Tracing

Large Language Models (LLMs) have demonstrated remarkable capabilities, yet their deployment is frequently undermined by undesirable behaviors such as generating harmful content, factual inaccuracies, and societal biases. Diagnosing the root causes of these failures poses a critical challenge for AI safety. Existing attribution methods, particularly those based on parameter gradients, often fall short due to prohibitive noisy signals and computational complexity. In this work, we introduce a novel and efficient framework that diagnoses a range of undesirable LLM behaviors by analyzing representation and its gradients, which operates directly in the model's activation space to provide a semantically meaningful signal linking outputs to their training data. We systematically evaluate our method for tasks that include tracking harmful content, detecting backdoor poisoning, and identifying knowledge contamination. The results demonstrate that our approach not only excels at sample-level attribution but also enables fine-grained token-level analysis, precisely identifying the specific samples and phrases that causally influence model behavior. This work provides a powerful diagnostic tool to understand, audit, and ultimately mitigate the risks associated with LLMs. The code is available at https://github.com/plumprc/RepT.

cs.CL↗