arXiv subjects

Zijing Li

Publications and source records attributed to Zijing Li.

7 recordsLinked to original sources

arXiv2026

TempJail: Temporal Jailbreak Attacks against Image-to-Video Generation Models

In recent years, image-to-video (I2V) generation models have made remarkable progress in subject consistency and temporal coherence, enabling high quality video synthesis. However, these advances also introduce new safety risks. Existing studies mainly focus on jailbreak attacks involving single frame violations, while largely overlooking the temporal dimension unique to video generation models. In this paper, we investigate three attack scenarios and uncover a temporal vulnerability in I2V systems: unsafe semantics may emerge not from a single frame, but from semantic composition over time. We further identify two key challenges in such attacks: temporal abstraction and semantic camouflage. To address these issues, we propose TempJail, a novel temporal jailbreak framework for I2V systems. For temporal abstraction, we decompose a target malicious caption into an initial frame visual condition and a temporal text instruction. For semantic camouflage, on the image side we model semantic injection as controlled latent perturbation in diffusion sampling and introduce gradient guidance from pretrained encoders. On the text side, we rewrite the caption into an innocuous ``subject-action-scene'' template that bypasses safety filters while preserving temporal guidance. In the black-box inference phase, these two modalities jointly enable malicious semantics to be gradually triggered over time. Experiments on closed-source commercial models, including Kling, Seedance, Veo and PixVerse, show that TempJail improves attack success rate over prior state-of-the-art methods by 23.3\% under GPT-5.2 evaluation and 22.0\% under human evaluation. Our codes are available at \href{https://github.com/luqi-glory/TempJail}{GitHub}.

cs.CV↗

arXiv2026

Imperceptible and Reversible Adversarial Examples against Vision-Language Models for Privacy Protection

Vision Language Models (VLMs) offer powerful multimodal ability but also expose users to text-based privacy attacks where adversaries crawl online photos and query VLMs to extract sensitive attributes. Existing reversible adversarial example (RAE) methods protect images in purely visual tasks but fail in multimodal settings, and current adversarial examples on VLMs rely on high frequency noise that severely degrades visual quality. We propose CloakDiff, the first framework for reversible, high fidelity privacy protection against text-based query attacks in VLMs. CloakDiff produces imperceptible adversarial examples by combining diffusion based adversarial editing with an invertible network that embeds the original image for lossless recovery. It perturbs both pixel space embeddings and manipulates latent cross attention maps to ensure strong cross-model and cross-prompt transferability while preserving global visual structure. To further enhance fidelity, we design EDM Heuristic Sampling, a principled diffusion schedule for adversarial guidance. Experiments on multiple datasets and VLMs demonstrate that CloakDiff delivers multimodal privacy preservation with high visual quality and reversibility.

cs.CV↗

arXiv2026

Synergistic improvement of specific strength and plasticity achieved in Ti-based metallic glass designed based on quasicrystal structure

Achieving a balance between low density, high strength, and good ductility remains a major challenge in the development of structural materials. Ti-based bulk metallic glasses (BMGs) have attracted considerable attention due to their exceptionally high specific strength. However, the intrinsic strength-plasticity trade-off has hindered their practical applications. Based on a quasicrystal-derived structural heredity and minor-element microalloying, this work realizes a synergistic enhancement of specific strength and plasticity in Ti-based BMGs. The resulting ((Ti_{40}Zr_{40}Ni_{20})_{72}Be_{28})_{97}Al_{3} BMGs demonstrate an ultrahigh specific strength of 5.34 \times 10^5 \text{ N}\cdot\text{m}\cdot\text{kg}^{-1}, establishing a new record for Ti-based BMGs, along with a plastic strain of 13\%, breaking through the traditional strength-plasticity limitation of BMGs. Structural analyses show that Al microalloying effectively inherits and modulates the short-range order derived from quasicrystalline structures, thereby achieving an observed synergistic enhancement in both strength and plasticity. This work provides new insights into composition design and lightweight structural applications of Ti-based BMGs.

cond-mat.mtrl-sci↗

arXiv2026

Efficient and Principled Scientific Discovery through Bayesian Optimization: A Tutorial

Traditional scientific discovery relies on an iterative hypothesise-experiment-refine cycle that has driven progress for centuries, but its intuitive, ad-hoc implementation often wastes resources, yields inefficient designs, and misses critical insights. This tutorial presents Bayesian Optimisation (BO), a principled probability-driven framework that formalises and automates this core scientific cycle. BO uses surrogate models (e.g., Gaussian processes) to model empirical observations as evolving hypotheses, and acquisition functions to guide experiment selection, balancing exploitation of known knowledge and exploration of uncharted domains to eliminate guesswork and manual trial-and-error. We first frame scientific discovery as an optimisation problem, then unpack BO's core components, end-to-end workflows, and real-world efficacy via case studies in catalysis, materials science, organic synthesis, and molecule discovery. We also cover critical technical extensions for scientific applications, including batched experimentation, heteroscedasticity, contextual optimisation, and human-in-the-loop integration. Tailored for a broad audience, this tutorial bridges AI advances in BO with practical natural science applications, offering tiered content to empower cross-disciplinary researchers to design more efficient experiments and accelerate principled scientific discovery.

cs.LG↗

arXiv2025

Bayesian Brain Edge-Based Connectivity (BBeC): a Bayesian model for brain edge-based connectivity inference

Brain connectivity analysis based on magnetic resonance imaging is crucial for understanding neurological mechanisms. However, edge-based connectivity inference faces significant challenges, particularly the curse of dimensionality when estimating high-dimensional covariance matrices. Existing methods often struggle to account for the unknown latent topological structure among brain edges, leading to inaccurate parameter estimation and unstable inference. To address these issues, this study proposes a Bayesian model based on a finite-dimensional Dirichlet distribution. Unlike non-parametric approaches, our method utilizes a finite-dimensional Dirichlet distribution to model the topological structure of brain networks, ensuring constant parameter dimensionality and improving algorithmic stability. We reformulate the covariance matrix structure to guarantee positive definiteness and employ a Metropolis-Hastings algorithm to simultaneously infer network topology and correlation parameters. Simulations validated the recovery of both network topology and correlation parameters. When applied to the Alzheimer's Disease Neuroimaging Initiative dataset, the model successfully identified structural subnetworks. The identified clusters were not only validated by composite anatomical metrics but also consistent with established findings in the literature, collectively demonstrating the model's reliability. The estimated covariance matrix also revealed that intragroup connection strength is stronger than intergroup connection strength. This study introduces a Bayesian framework for inferring brain network topology and high-dimensional covariance structures. The model configuration reduces parameter dimensionality while ensuring the positive definiteness of covariance matrices. As a result, it offers a reliable tool for investigating intrinsic brain connectivity in large-scale neuroimaging studies.

stat.ME↗

arXiv2025

Sparse Broad Learning System via Sequential Threshold Least-Squares for Nonlinear System Identification under Noise

The Broad Learning System (BLS) has gained significant attention for its computational efficiency and less network parameters compared to deep learning structures. However, the standard BLS relies on the pseudoinverse solution, which minimizes the mean square error with $L_2$-norm but lacks robustness against sensor noise and outliers common in industrial environments. To address this limitation, this paper proposes a novel Sparse Broad Learning System (S-BLS) framework. Instead of the traditional ridge regression, we incorporate the Sequential Threshold Least-Squares (STLS) algorithm -- originally utilized in the sparse identification of nonlinear dynamics (SINDy) -- into the output weight learning process of BLS. By iteratively thresholding small coefficients, the proposed method promotes sparsity in the output weights, effectively filtering out noise components while maintaining modeling accuracy. This approach falls under the category of sparse regression and is particularly suitable for noisy environments. Experimental results on a numerical nonlinear system and a noisy Continuous Stirred Tank Reactor (CSTR) benchmark demonstrate that the proposed method is effective and achieves superior robustness compared to standard BLS.

eess.SY↗

arXiv2025

Vanish into Thin Air: Cross-prompt Universal Adversarial Attacks for SAM2

Recent studies reveal the vulnerability of the image segmentation foundation model SAM to adversarial examples. Its successor, SAM2, has attracted significant attention due to its strong generalization capability in video segmentation. However, its robustness remains unexplored, and it is unclear whether existing attacks on SAM can be directly transferred to SAM2. In this paper, we first analyze the performance gap of existing attacks between SAM and SAM2 and highlight two key challenges arising from their architectural differences: directional guidance from the prompt and semantic entanglement across consecutive frames. To address these issues, we propose UAP-SAM2, the first cross-prompt universal adversarial attack against SAM2 driven by dual semantic deviation. For cross-prompt transferability, we begin by designing a target-scanning strategy that divides each frame into k regions, each randomly assigned a prompt, to reduce prompt dependency during optimization. For effectiveness, we design a dual semantic deviation framework that optimizes a UAP by distorting the semantics within the current frame and disrupting the semantic consistency across consecutive frames. Extensive experiments on six datasets across two segmentation tasks demonstrate the effectiveness of the proposed method for SAM2. The comparative results show that UAP-SAM2 significantly outperforms state-of-the-art (SOTA) attacks by a large margin.

cs.CV↗