Searcharxiv⌕ Search

arXiv subjects

Zelin Zhang

Publications and source records attributed to Zelin Zhang.

At least 19 recordsLinked to original sources

GapLeak: When Privacy-Preserving LLM-based Web Agents Become a Privacy Risk

Large Language Model (LLM)-based web agents rely on data minimization and least privilege to protect user privacy. We show that these principles are necessary but insufficient: they constrain authorized sources, but not higher-sensitivity sources that remain reachable. A website can request a field unavailable from authorized sources, creating an information gap. To complete the task, a goal-driven agent may escalate to reachable but unauthorized sources and disclose accumulated sensitive information. We formalize this vulnerability as GapLeak, where an adversary exploits a legitimate web form without prompt injection or other malicious instructions. We evaluate GapLeak on three web-agent frameworks and six LLMs from Anthropic, Google, and OpenAI across recruitment, e-commerce, travel, and healthcare. GapLeak induces an average unauthorized disclosure of approximately 37%. The information gap is the primary driver: eliminating it reduces disclosure by more than 50%. Existing instruction-level defenses based on least privilege partially mitigate the vulnerability but often reduce task success. We propose submission-time verification, which filters unauthorized content before form submission, eliminating over 96% of unauthorized disclosures while preserving task success.

cs.CR↗

Gravitational dynamics and backward ray-traced shadows of bumblebee AdS black holes

We investigate the total observed frequency shift of photons emitted by massive geodesic particles in circular orbits near the event horizon of bumblebee AdS black holes. The cosmological effects of the bumblebee field are investigated through dynamical equations governing the evolution and characteristic features of the Universe. In addition, the bumblebee field generates cosmological anisotropies by acting as a source of anisotropy and selecting a preferred spatial direction. We formulate the characteristic equations for the innermost stable circular orbit using the effective potential approach. Further analysis is devoted to establishing a functional relationship between the BH mass and the observed redshift and blueshift of photons emitted from the accretion disk. Finally, we study the shadow cast by the bumblebee AdS BH using the backward ray tracing process implemented in the Arcmancer code, where photon geodesics are initialized at the observers location and integrated backward in time to generate BH images, typically illuminated by electromagnetic radiation from nearby accretion flows.

gr-qc↗

EventSeeColor: Learning to Reconstruct Color from Monochrome Events

Event cameras capture visual brightness changes under challenging lighting conditions. However, traditional event-based image reconstruction targets grayscale intensity and produces images without scene color. Color distinguishes surfaces with similar intensity. Reconstructing color from events can reveal appearance differences that reconstructed grayscale images cannot express. Although monochrome event cameras do not directly measure color, their event-triggering patterns depend on surface reflectance, illumination, and the sensor's spectral response. These patterns may contain cues that support color reconstruction. In this paper, we present EventSeeColor, which uses event-reconstructed grayscale and aligned events to jointly guide a diffusion colorizer. Grayscale-derived semantic context supports color inference, while region-balanced chroma supervision and local RGB refinement improve color distribution, brightness, and texture. In the combined evaluation on two real event datasets, EventSeeColor achieves the best structural similarity, learned perceptual scores, and chromatic-region color difference among the external methods.

cs.CV↗

Are Watermarked Images Editable? SafeMark for Watermark-Preserving Text-Guided Image Editing

This paper investigates a fundamental yet underexplored question: can watermarked images remain editable without compromising watermark integrity? We propose SafeMark, a framework for watermark-preserving text-guided image manipulation that explicitly integrates watermark integrity into the editing process. Specifically, SafeMark adds a thresholded watermark-decoding loss directly to the diffusion editor's training objective, fine-tuning the editor so that semantically valid edits also preserve the embedded watermark at the final output. This design admits a clean information-theoretic justification: maintaining high bit-accuracy on the edited image lower-bounds the mutual information that the editor channel preserves between watermark and edited output, the quantity that fundamentally controls watermark recoverability. SafeMark is compatible with differentiable diffusion-based editors, and requires no architectural modification. Extensive evaluations across multiple datasets, text-guided editing methods, and post-edit distortion settings demonstrate that SafeMark achieves high watermark bit accuracy across diverse editing settings while maintaining high-quality semantic edits, without sacrificing robustness to common post-edit distortions. These results demonstrate that semantic editability and watermark integrity are fundamentally compatible, enabling trustworthy image provenance in generative editing pipelines.

cs.CV↗

WaveMark: Robust Wavelet-Domain Image Watermarking for Diffusion Models

Image watermarking enables provenance tracing for photorealistic images generated by diffusion models. However, state-of-the-art (SOTA) removal attacks can suppress embedded signals with little perceptual degradation. This vulnerability limits the effectiveness of watermarks for intellectual property protection and misuse attribution. We propose WaveMark, a novel wavelet-domain watermarking method that embeds messages in a diffusion model's initial noise. To support robust recovery from perturbed carriers, WaveMark distributes keyed replicas of message bits across spatial positions, wavelet orientations, and scales. Each carrier groups coefficients from four latent channels and encodes a randomized bit through magnitude projection, providing an explicit decoding margin. We evaluate WaveMark with SD2.1 on SDP and MS-COCO Captions under 14 attacks. WaveMark achieves 100.0% bit accuracy (BA) without attacks, while maintaining image quality comparable to the evaluated baselines. Under SOTA MarkNull and MarkNull-A, it retains over 88% BA, outperforming all baselines.

cs.CR↗

Autocorrelation in black hole flare movies: Departures from Kerr

Autocorrelation in black hole movies encodes the time delays and angular separations between gravitationally lensed images, offering a means of testing the spacetime geometry. Using an orbiting hot spot in the parametrized Johannsen-Psaltis metric with a single deformation parameter, we investigate the ability of lensing-induced correlations to distinguish departures from Kerr. From ray-traced movies, we extract the slope (k) and intercept (b) of an approximately linear secondary ridge in the time-angle autocorrelation map as potential future observables. Variations in the spin and the metric deformation parameter produce distinctive trajectories in the (k,b) plane, allowing the two observables to jointly constrain both parameters when the orbital radius, observer inclination, and orbital angular velocity are specified. By varying the orbital angular velocity, we construct a Kerr reference region in the (k,b) plane, and we find that, for some low-spin configurations with negative deformation parameters, the measured slope-intercept pairs of the secondary peaks lie outside this reference region. This feature appears for sub-Keplerian, Keplerian, and super-Keplerian motion, indicating that secondary-ridge measurements from black hole flare movies have the potential to distinguish non-Kerr signatures.

astro-ph.HE↗

KPolaris: GPU-accelerated Polarized Radiative Transfer in General Relativity

Polarized black-hole images carry information about both the magnetized plasma and the curved spacetime through which the radiation propagates. We present KPolaris, an open-source GPU-accelerated general-relativistic polarized radiative-transfer code built on the Kokkos framework. It supports analytic plasma models and GRMHD simulation data, with both fast-light and slow-light radiative transfer. The code also provides diagnostics that connect image structure to the emitting plasma and to propagation effects along the rays. These include emissivity-weighted plasma maps, emission contributions from selected regions, selected-ray histories, and sensitivities of the Stokes parameters to plasma parameters. Numerical convergence tests and comparisons with an independent radiative-transfer code assess the accuracy of the calculations. GPU acceleration reduces the computational cost of image generation, facilitating multi-frequency and time-dependent studies of polarized black-hole images.

astro-ph.HE↗

Dense Dump Codec: Error-Controlled, Random-Access Compression of GRMHD Time Series for Slow-Light Radiative Transfer

Black-hole movies provide a unique venue to study the time evolution of accreting plasma. Modeling this evolution with slow-light radiative transfer requires closely spaced simulation outputs, because the plasma changes as light propagates through it. Saving these outputs in full is costly, whereas sparse sampling introduces time-interpolation errors. We present the Dense Dump Codec (DDC), a compression method for dense GRMHD time series. DDC stores exact anchor states and compactly represents the intermediate evolution, while allowing direct access to selected times and variables. Applied to eight production sequences spanning SANE and MAD simulations at several black-hole spins with 0.1M output, DDC archives are about 10 times smaller than the original dense data and even about 50% smaller than the original 0.5M output in aggregate. We further integrate a native DDC reader into polarized slow-light radiative transfer. Tests at 86, 230, and 345 GHz show that DDC may reduce aggregate Stokes-image errors by 54-79% relative to slow-light calculations using original 0.5M input. These results indicate that DDC enables high-cadence slow-light calculations without the prohibitive cost of saving every dense GRMHD state in its original form.

astro-ph.HE↗

Thermodynamics of Kerr-Newman-Bertotti-Robinson black holes

In this work, we extend the thermodynamic analyses of the neutral and specially charged Kerr-Bertotti-Robinson black holes to the general Kerr-Newman-Bertotti-Robinson family, in which the electric and external-field parameters are independent. Using covariant surface charges and canonical integrability methods, we determine the total angular momentum and the canonical mass. The angular momentum follows analytically from the combined gravitational and electromagnetic surface charges. However, the infinitesimal charge associated with coordinate time translations is not integrable in solution space, so the mass must be associated with a more general symmetry generator. Imposing the canonical integrability conditions on this generator, together with the Kerr-Newman mass as the zero-field boundary condition, selects the Christodoulou-Ruffini mass and determines the associated thermodynamic potentials. The first law and Smarr formula take the same form as the ones in usual Kerr-Newman case, and the two previously studied Kerr-Bertotti-Robinson cases are recovered as special limits.

gr-qc↗

Shadows and Thin-Disk Images of Kerr-Newman Black Holes in a Bertotti-Robinson Magnetic Field

In this paper, we investigate the optical properties of Kerr-Newman-Bertotti-Robinson (KN-BR) black holes. We use the separability of null geodesics to analyze unstable spherical photon orbits and determine the radial extent of the photon shell. Because the spacetime is not asymptotically flat, we construct the critical curve on the screen of a finite-distance zero-angular-momentum observer. We then perform backward ray tracing for a geometrically thin and optically thin disk that extends from the outer region to the event horizon, and examine the resulting images, intensity profiles, critical-curve areas, and inner-shadow areas. We find that the genuine neutral Kerr-BR$_0$ and specially charged Kerr-BR$_s$ configurations have nearly identical optical appearances. It is remarkable that for the KN-BR black holes increasing the electric charge reduces the characteristic image size in the Kerr-Newman limit but enlarges it in the magnetized configurations considered here. We also find that the external magnetic field strongly increases the apparent image scale, while the observer inclination affects the inner-shadow area more significantly than the critical-curve area. These results may provide useful theoretical insight for future observations aimed at identifying such exotic magnetized black holes.

gr-qc↗

MarkNull: Model-Agnostic Watermark Removal in AI-Generated Images via On-Manifold Latent Manipulation

Digital watermarking has emerged as a critical technique for provenance and copyright attribution in AI-generated imagery, yet its robustness against realistic, model-agnostic removal attacks remains poorly explored. Existing attacks either succeed only against specific generative models or achieve removal at the cost of severe visual degradation. In this paper, we propose MarkNull, a model-agnostic watermark removal attack via on-manifold latent manipulation. MarkNull is grounded in a key observation: watermarked images exhibit a strong statistical dependency between the generated latent representation and the embedded initial noise. To quantify this dependency, we introduce the Noise-Latent Alignment Score (NLAS) and formulate an optimization objective that selectively decorrelates the latent representation from the embedded watermark while preserving semantic fidelity. Extensive evaluations across different categories of watermarking paradigms, including post-hoc, fine-tuning-based, and initial-noise-based schemes, demonstrate that MarkNull reduces average bit accuracy to 53.14%, approaching random-guessing (50%), without perceptible image degradation. To further improve scalability, we propose MarkNull-A, an amortized, optimization-free variant that distills the attack into a single forward pass, achieving 0.50 s/image with modest computational overhead. Notably, our attacks successfully compromise Google's SynthID-Image system while preserving high visual quality and transfer effectively to video watermarking. Finally, we present an attack detection mechanism as a defensive counterpart to MarkNull and MarkNull-A, highlighting the necessity of developing watermark designs resilient to model-agnostic latent-space attacks.

cs.CR↗

LoginTrap: Uncovering Task-Agnostic Phishing-Style Indirect Prompt Injection Attacks against LLM-based Web Agents

LLM-based web agents automate user tasks by observing webpages and executing browser actions on behalf of users. As these agents operate on real web services, login becomes a sensitive authentication boundary because it involves credentials and sensitive information. Existing work shows that malicious webpage content can manipulate web agent actions, but it has not fully examined whether such content can induce login and cause end-to-end private data leakage. We study this attack surface and present LoginTrap, a task-agnostic login-inducing attack against LLM-based web agents. LoginTrap assumes a black box attacker that controls the webpage context and the induced login flow without knowing the user task or web agent internals. Under this threat model, LoginTrap uses webpage context to generate page-specific indirect injections through a fuzzing-inspired process, making login appear as a plausible prerequisite for continuing the task and guiding the agent to a controlled login page. We conduct a comprehensive analysis of LoginTrap across realistic web agent executions. The results show that LoginTrap reaches 86\% average end-to-end attack success across LLM backbones and remains effective across agent architectures and defenses. These findings identify login inducement as a systematic authentication boundary risk and motivate further research on authentication-aware defenses for web agents.

cs.CR↗

CrossWeaver: Cross-modal Weaving for Arbitrary-Modality Semantic Segmentation

Multimodal semantic segmentation has shown great potential in leveraging complementary information across diverse sensing modalities. However, existing approaches often rely on carefully designed fusion strategies that either use modality-specific adaptations or rely on loosely coupled interactions, thereby limiting flexibility and resulting in less effective cross-modal coordination. Moreover, these methods often struggle to balance efficient information exchange with preserving the unique characteristics of each modality across different modality combinations. To address these challenges, we propose CrossWeaver, a simple yet effective multimodal fusion framework for arbitrary-modality semantic segmentation. Its core is a Modality Interaction Block (MIB), which enables selective and reliability-aware cross-modal interaction within the encoder, while a lightweight Seam-Aligned Fusion (SAF) module further aggregates the enhanced features. Extensive experiments on multiple multimodal semantic segmentation benchmarks demonstrate that our framework achieves state-of-the-art performance with minimal additional parameters and strong generalization to unseen modality combinations.

cs.CV↗

Images of Braneworld black holes with radiatively inefficient accretion flows

Horizon-scale imaging acts as a transformative tool for probing spacetime geometry, enabling stringent tests of gravitational theories in the strong-field regime. The Casadio-Fabbri-Mazzacurati(CFM) black hole in braneworld contains an extra parameter that characterizes the tidal effects from the bulk geometry, making it highly valuable for this task. We perform general relativistic radiative transfer (GRRT) simulations and generate synthetic images consistent with Event Horizon Telescope observations of M87*. We find that the tidal parameter imprints nonmonotonic changes on the image morphology, underscoring the intricate coupling between spacetime geometry and the observable radiation from the accreting plasma. We also analyze the image-comparison metric using normalized cross-correlation coefficients and the DSSIM index and find that the magnitudes of these mismatches are on the order of 10^3, which implies that identifying braneworld black holes through black hole images remains challenging even with future ngEHT and BHEX observations.

gr-qc↗

Security, Privacy, and Ethical Risks in OpenClaw

This paper systematically investigates the security, privacy, and ethical risks, as well as the traceability challenges of OpenClaw, a locally executable AI agent system for natural language interaction and real-world task completion. While OpenClaw shows strong potential for personal assistance, office automation, cross-platform task management, and information integration, it also raises serious security, privacy, and ethical concerns. By analyzing its system architecture, core functionalities, deployment model, and representative application scenarios, this paper aims to reveal the risks that may arise when such a highly privileged agent is integrated into personal and organizational digital environments. We focus in particular on the challenges associated with persistent local storage, tool invocation, cross-context information aggregation, multi-user interaction, and the integration of plugins and external services. We argue that these issues constitute major barriers to the trustworthy deployment and widespread adoption of this technology. Finally, we summarize the open challenges in security defenses, privacy protection, ethical governance, and traceability in agent use, and call for joint efforts from researchers, developers, deployers, and regulators to build AI agent systems that are safer, more reliable, and more trustworthy.

cs.CR↗

From AI-Generated Content to Agentic Action: Security and Safety Threats in Generative AI

Generative AI systems are increasingly used not only to produce content but also to retrieve data, invoke tools, and execute actions. This work examines the security and safety implications of that shift across content-level, model-level, and agentic threats. We analyze how attacker access requirements, system autonomy, and the scope of potential harm change as models move from generating artifacts to executing operations through tool chains and external APIs. We then assess technical countermeasures including detection, watermarking, alignment, and emerging agentic safeguards, and show that several depend on forms of institutional coordination that current governance arrangements do not yet provide. Across the cases examined, capability deployment and attack-surface expansion repeatedly outpace defensive responses as systems move from generating content to executing real-world actions.

cs.CR↗

When and Why Does Unsupervised RL Succeed in Mathematical Reasoning? A Manifold Envelopment Perspective

Although outcome-based reinforcement learning (RL) significantly advances the mathematical reasoning capabilities of Large Language Models (LLMs), its reliance on computationally expensive ground-truth annotations imposes a severe scalability bottleneck. Unsupervised RL guided by intrinsic rewards offers a scalable alternative, yet it suffers from opaque training dynamics and catastrophic instability, such as policy collapse and reward hacking. In this paper, we first design and evaluate a suite of intrinsic rewards that explicitly enforce concise and certain generation. Second, to discover the boundaries of this approach, we test base models across a spectrum of intrinsic reasoning capabilities, revealing how a model's foundational logical prior dictates its success or failure. Finally, to demystify why certain configurations stabilize while others collapse, we introduce a novel geometric diagnostic lens, showing that successful cases are enveloped by manifolds. Ultimately, our work goes beyond merely demonstrating that enforcing concise and certain responses successfully boosts mathematical reasoning; we reveal when this unsupervised approach breaks down and geometrically diagnose why.

cs.LG↗

Secure and Robust Watermarking for AI-generated Images: A Comprehensive Survey

The rapid progress of Generative Artificial Intelligence (GenAI) has enabled the effortless synthesis of high-quality visual content, while simultaneously raising pressing concerns about intellectual property protection, authenticity, and accountability. Among various countermeasures, watermarking has emerged as a fundamental mechanism for tracing provenance, distinguishing AI-generated images from natural content, and supporting trustworthy digital ecosystems. This paper presents a comprehensive survey of AI-generated image watermarking, systematically reviewing the field from five perspectives: (1) the formalization and fundamental components of image watermarking systems; (2) existing watermarking methodologies and their comparative characteristics; (3) evaluation metrics in terms of visual fidelity, embedding capacity, and detectability; (4) known vulnerabilities under malicious attacks and recent advances in secure and robust watermarking designs; and (5) open challenges, emerging trends, and future research directions. The survey seeks to offer researchers a holistic understanding of watermarking technologies for AI-generated images and to facilitate their continued advancement toward secure and responsible AI-generated content practices.

cs.CR↗