Proactive watermarking offers a promising approach for deepfake tamper detection and localization in short-form videos. However, existing methods often decouple audio and visual evidence and assume that watermark signals remain reliable under real-world degradations, making tamper localization vulnerable to multimodal misalignment and compression distortions. Moreover, existing semi-fragile visual watermarking methods often degrade significantly under codec compression because their embedding bands overlap with compression-sensitive frequency regions. To address these limitations, we propose Layered Audio-Visual Anti-tampering Watermarking (LAVA), a calibration-aware audio-visual watermark fusion framework for deepfake tamper detection and localization. LAVA leverages cross-modal watermark fusion and calibration-aware alignment to preserve consistent and reliable tamper evidence under compression and audio-visual asynchrony, enabling robust tamper localization. Extensive experiments demonstrate that LAVA achieves near-perfect detection performance (AP = 0.999), remains robust to compression and multimodal misalignment, and significantly improves tamper localization reliability over existing audio-visual fusion baselines.

LLM-based Multi-Agent Systems (MAS) have demonstrated remarkable capabilities in solving complex tasks. Central to MAS is the communication topology which governs how agents exchange information internally. Consequently, the security of communication topologies has attracted increasing attention. In this paper, we investigate a critical privacy risk: MAS communication topologies can be inferred under a restrictive black-box setting, exposing system vulnerabilities and posing significant intellectual property threats. To explore this risk, we propose Communication Inference Attack (CIA), a novel attack that constructs new adversarial queries to induce intermediate agents' reasoning outputs and models their semantic correlations through the proposed global bias disentanglement and LLM-guided weak supervision. Extensive experiments on MAS with optimized communication topologies demonstrate the effectiveness of CIA, achieving an average AUC of 0.87 and a peak AUC of up to 0.99, thereby revealing the substantial privacy risk in MAS.

LLM-based coding agents extend their capabilities via third-party agent skills distributed through open marketplaces without mandatory security review. Unlike traditional packages, these skills are executed as operational directives with system-level privileges, so a single malicious skill can compromise the host. Prior work has not examined whether supply-chain attacks can directly hijack an agent's action space, such as file writes, shell commands, and network requests, despite existing safeguards. We introduce Document-Driven Implicit Payload Execution (DDIPE), which embeds malicious logic in code examples and configuration templates within skill documentation. Because agents reuse these examples during normal tasks, the payload executes without explicit prompts. Using an LLM-driven pipeline, we generate 1,070 adversarial skills from 81 seeds across 15 MITRE ATTACK categories. Across four frameworks and five models, DDIPE achieves 11.6% to 33.5% bypass rates, while explicit instruction attacks achieve 0% under strong defenses. Static analysis detects most cases, but 2.5% evade both detection and alignment. Responsible disclosure led to four confirmed vulnerabilities and two fixes.

Third-party skills extend LLM agents with powerful capabilities but often handle sensitive credentials in privileged environments, making leakage risks poorly understood. We present the first large-scale empirical study of this problem, analyzing 17,022 skills (sampled from 170,226 on SkillsMP) using static analysis, sandbox testing, and manual inspection. We identify 520 vulnerable skills with 1,708 issues and derive a taxonomy of 10 leakage patterns (4 accidental and 6 adversarial). We find that (1) leakage is fundamentally cross-modal: 76.3% require joint analysis of code and natural language, while 3.1% arise purely from prompt injection; (2) debug logging is the primary vector, with print and console.log causing 73.5% of leaks due to stdout exposure to LLMs; and (3) leaked credentials are both exploitable (89.6% without privileges) and persistent, as forks retain secrets even after upstream fixes. After disclosure, all malicious skills were removed and 91.6% of hardcoded credentials were fixed. We release our dataset, taxonomy, and detection pipeline to support future research.

Diffusion-based watermarking methods embed verifiable marks by manipulating the initial noise or the reverse diffusion trajectory. However, these methods share a critical assumption: verification can succeed only if the diffusion trajectory can be faithfully reconstructed. This reliance on trajectory recovery constitutes a fundamental and exploitable vulnerability. We propose $\underline{\mathbf{S}}$tochastic $\underline{\mathbf{Hi}}$dden-Trajectory De$\underline{\mathbf{f}}$lec$\underline{\mathbf{t}}$ion ($\mathbf{SHIFT}$), a training-free attack that exploits this common weakness across diverse watermarking paradigms. SHIFT leverages stochastic diffusion resampling to deflect the generative trajectory in latent space, making the reconstructed image statistically decoupled from the original watermark-embedded trajectory while preserving strong visual quality and semantic consistency. Extensive experiments on nine representative watermarking methods spanning noise-space, frequency-domain, and optimization-based paradigms show that SHIFT achieves 95%--100% attack success rates with nearly no loss in semantic quality, without requiring any watermark-specific knowledge or model retraining.

Federated Learning (FL) enables collaborative model training by sharing model updates instead of raw data, aiming to protect user privacy. However, recent studies reveal that these shared updates can inadvertently leak sensitive training data through gradient inversion attacks (GIAs). Among them, active GIAs are particularly powerful, enabling high-fidelity reconstruction of individual samples even under large batch sizes. Nevertheless, existing approaches often require architectural modifications, which limit their practical applicability. In this work, we bridge this gap by introducing the Activation REcovery via Sparse inversion (ARES) attack, an active GIA designed to reconstruct training samples from large training batches without requiring architectural modifications. Specifically, we formulate the recovery problem as a noisy sparse recovery task and solve it using the generalized Least Absolute Shrinkage and Selection Operator (Lasso). To extend the attack to multi-sample recovery, ARES incorporates the imprint method to disentangle activations, enabling scalable per-sample reconstruction. We further establish the expected recovery rate and derive an upper bound on the reconstruction error, providing theoretical guarantees for the ARES attack. Extensive experiments on CNNs and MLPs demonstrate that ARES achieves high-fidelity reconstruction across diverse datasets, significantly outperforming prior GIAs under large batch sizes and realistic FL settings. Our results highlight that intermediate activations pose a serious and underestimated privacy risk in FL, underscoring the urgent need for stronger defenses.

Semantic segmentation models are widely deployed in safety-critical applications such as autonomous driving, yet their vulnerability to backdoor attacks remains largely underexplored. Prior segmentation backdoor studies transfer threat settings from existing image classification tasks, focusing primarily on object-to-background mis-segmentation. In this work, we revisit the threats by systematically examining backdoor attacks tailored to semantic segmentation. We identify four coarse-grained attack vectors (Object-to-Object, Object-to-Background, Background-to-Object, and Background-to-Background attacks), as well as two fine-grained vectors (Instance-Level and Conditional attacks). To formalize these attacks, we introduce BADSEG, a unified framework that optimizes trigger designs and applies label manipulation strategies to maximize attack performance while preserving victim model utility. Extensive experiments across diverse segmentation architectures on benchmark datasets demonstrate that BADSEG achieves high attack effectiveness with minimal impact on clean samples. We further evaluate six representative defenses and find that they fail to reliably mitigate our attacks, revealing critical gaps in current defenses. Finally, we demonstrate that these vulnerabilities persist in recent emerging architectures, including transformer-based networks and the Segment Anything Model (SAM), thereby compromising their security. Our work reveals previously overlooked security vulnerabilities in semantic segmentation, and motivates the development of defenses tailored to segmentation-specific threat models.

Watermarking the initial noise of diffusion models has emerged as a promising approach for image provenance, but content-independent noise patterns can be forged via inversion and regeneration attacks. Recent semantic-aware watermarking methods improve robustness by conditioning verification on image semantics. However, their reliance on a single global semantic binding makes them vulnerable to localized but globally coherent semantic edits. To address this limitation and provide a trustworthy semantic-aware watermark, we propose $\underline{\textbf{S}}$emantic $\underline{\textbf{L}}$atent $\underline{\textbf{I}}$njection via $\underline{\textbf{C}}$ompartmentalized $\underline{\textbf{E}}$mbedding ($\textbf{SLICE}$). Our framework decouples image semantics into four semantic factors (subject, environment, action, and detail) and precisely anchors them to distinct regions in the initial Gaussian noise. This fine-grained semantic binding enables advanced watermark verification where semantic tampering is detectable and localizable. We theoretically justify why SLICE enables robust and reliable tamper localization and provides statistical guarantees on false-accept rates. Experimental results demonstrate that SLICE significantly outperforms existing baselines against advanced semantic-guided regeneration attacks, substantially reducing attack success while preserving image quality and semantic fidelity. Overall, SLICE offers a practical, training-free provenance solution that is both fine-grained in diagnosis and robust to realistic adversarial manipulations.

Generative images have proliferated on Web platforms in social media and online copyright distribution scenarios, and semantic watermarking has increasingly been integrated into diffusion models to support reliable provenance tracking and forgery prevention for web content. Traditional noise-layer-based watermarking, however, remains vulnerable to inversion attacks that can recover embedded signals. To mitigate this, recent content-aware semantic watermarking schemes bind watermark signals to high-level image semantics, constraining local edits that would otherwise disrupt global coherence. Yet, large language models (LLMs) possess structured reasoning capabilities that enable targeted exploration of semantic spaces, allowing locally fine-grained but globally coherent semantic alterations that invalidate such bindings. To expose this overlooked vulnerability, we introduce a Coherence-Preserving Semantic Injection (CSI) attack that leverages LLM-guided semantic manipulation under embedding-space similarity constraints. This alignment enforces visual-semantic consistency while selectively perturbing watermark-relevant semantics, ultimately inducing detector misclassification. Extensive empirical results show that CSI consistently outperforms prevailing attack baselines against content-aware semantic watermarking, revealing a fundamental security weakness of current semantic watermark designs when confronted with LLM-driven semantic perturbations.

Code models are increasingly adopted in software development but remain vulnerable to backdoor attacks via poisoned training data. Existing backdoor attacks on code models face a fundamental trade-off between transferability and stealthiness. Static trigger-based attacks insert fixed dead code patterns that transfer well across models and datasets but are easily detected by code-specific defenses. In contrast, dynamic trigger-based attacks adaptively generate context-aware triggers to evade detection but suffer from poor cross-dataset transferability. Moreover, they rely on unrealistic assumptions of identical data distributions between poisoned and victim training data, limiting their practicality. To overcome these limitations, we propose Sharpness-aware Transferable Adversarial Backdoor (STAB), a novel attack that achieves both transferability and stealthiness without requiring complete victim data. STAB is motivated by the observation that adversarial perturbations in flat regions of the loss landscape transfer more effectively across datasets than those in sharp minima. To this end, we train a surrogate model using Sharpness-Aware Minimization to guide model parameters toward flat loss regions, and employ Gumbel-Softmax optimization to enable differentiable search over discrete trigger tokens for generating context-aware adversarial triggers. Experiments across three datasets and two code models show that STAB outperforms prior attacks in terms of transferability and stealthiness. It achieves a 73.2% average attack success rate after defense, outperforming static trigger-based attacks that fail under defense. STAB also surpasses the best dynamic trigger-based attack by 12.4% in cross-dataset attack success rate and maintains performance on clean inputs.

Third-party agent skills extend LLM-based agents with instruction files and executable code that run on users' machines. Skills execute with user privileges and are distributed through community registries with minimal vetting, but no ground-truth dataset exists to characterize the resulting threats. We construct the first labeled dataset of malicious agent skills by behaviorally verifying 98,380 skills from two community registries, confirming 157 malicious skills with 632 vulnerabilities. These attacks are not incidental. Malicious skills average 4.03 vulnerabilities across a median of three kill chain phases, and the ecosystem has split into two archetypes: Data Thieves that exfiltrate credentials through supply chain techniques, and Agent Hijackers that subvert agent decision-making through instruction manipulation. A single actor accounts for 54.1\% of confirmed cases through templated brand impersonation. Shadow features, capabilities absent from public documentation, appear in 0\% of basic attacks but 100\% of advanced ones; several skills go further by exploiting the AI platform's own hook system and permission flags. Responsible disclosure led to 93.6\% removal within 30 days. We release the dataset and analysis pipeline to support future work on agent skill security.

Interpretable time series deep learning systems are often assessed by checking temporal consistency on explanations, implicitly treating this as evidence of robustness. We show that this assumption can fail: Predictions and explanations can be adversarially decoupled, enabling targeted misclassification while the explanation remains plausible and consistent with a chosen reference rationale. We propose TSEF (Time Series Explanation Fooler), a dual-target attack that jointly manipulates the classifier and explainer outputs. In contrast to single-objective misclassification attacks that disrupt explanation and spread attribution mass broadly, TSEF achieves targeted prediction changes while keeping explanations consistent with the reference. Across multiple datasets and explainer backbones, our results consistently reveal that explanation stability is a misleading proxy for decision robustness and motivate coupling-aware robustness evaluations for trustworthy time series tasks.

Machine unlearning is an emerging technique that aims to remove the influence of specific data from trained models, thereby enhancing privacy protection. However, recent research has uncovered critical privacy vulnerabilities, showing that adversaries can exploit unlearning inversion to reconstruct data that was intended to be erased. Despite the severity of this threat, dedicated defenses remain lacking. To address this gap, we propose UnlearnShield, the first defense specifically tailored to counter unlearning inversion. UnlearnShield introduces directional perturbations in the cosine representation space and regulates them through a constraint module to jointly preserve model accuracy and forgetting efficacy, thereby reducing inversion risk while maintaining utility. Experiments demonstrate that it achieves a good trade-off among privacy protection, accuracy, and forgetting.

Federated Rank Learning (FRL) is a promising Federated Learning (FL) paradigm designed to be resilient against model poisoning attacks due to its discrete, ranking-based update mechanism. Unlike traditional FL methods that rely on model updates, FRL leverages discrete rankings as a communication parameter between clients and the server. This approach significantly reduces communication costs and limits an adversary's ability to scale or optimize malicious updates in the continuous space, thereby enhancing its robustness. This makes FRL particularly appealing for applications where system security and data privacy are crucial, such as web-based auction and bidding platforms. While FRL substantially reduces the attack surface, we demonstrate that it remains vulnerable to a new class of local model poisoning attack, i.e., fine-grained control attacks. We introduce the Edge Control Attack (ECA), the first fine-grained control attack tailored to ranking-based FL frameworks. Unlike conventional denial-of-service (DoS) attacks that cause conspicuous disruptions, ECA enables an adversary to precisely degrade a competitor's accuracy to any target level while maintaining a normal-looking convergence trajectory, thereby avoiding detection. ECA operates in two stages: (i) identifying and manipulating Ascending and Descending Edges to align the global model with the target model, and (ii) widening the selection boundary gap to stabilize the global model at the target accuracy. Extensive experiments across seven benchmark datasets and nine Byzantine-robust aggregation rules (AGRs) show that ECA achieves fine-grained accuracy control with an average error of only 0.224%, outperforming the baseline by up to 17x. Our findings highlight the need for stronger defenses against advanced poisoning attacks. Our code is available at: https://github.com/Chenzh0205/ECA

Existing segmentation models exhibit significant vulnerability to adversarial attacks.To improve robustness, adversarial training incorporates adversarial examples into model training. However, existing attack methods consider only global semantic information and ignore contextual semantic relationships within the samples, limiting the effectiveness of adversarial training. To address this issue, we propose EroSeg-AT, a vulnerability-aware adversarial training framework that leverages EroSeg to generate adversarial examples. EroSeg first selects sensitive pixels based on pixel-level confidence and then progressively propagates perturbations to higher-confidence pixels, effectively disrupting the semantic consistency of the samples. Experimental results show that, compared to existing methods, our approach significantly improves attack effectiveness and enhances model robustness under adversarial training.

Hard-label black-box settings, where only top-1 predicted labels are observable, pose a fundamentally constrained yet practically important feedback model for understanding model behavior. A central challenge in this regime is whether meaningful gradient information can be recovered from such discrete responses. In this work, we develop a unified theoretical perspective showing that a wide range of existing sign-flipping hard-label attacks can be interpreted as implicitly approximating the sign of the true loss gradient. This observation reframes hard-label attacks from heuristic search procedures into instances of gradient sign recovery under extremely limited feedback. Motivated by this first-principles understanding, we propose a new attack framework that combines a zero-query frequency-domain initialization with a Pattern-Driven Optimization (PDO) strategy. We establish theoretical guarantees demonstrating that, under mild assumptions, our initialization achieves higher expected cosine similarity to the true gradient sign compared to random baselines, while the proposed PDO procedure attains substantially lower query complexity than existing structured search approaches. We empirically validate our framework through extensive experiments on CIFAR-10, ImageNet, and ObjectNet, covering standard and adversarially trained models, commercial APIs, and CLIP-based models. The results show that our method consistently surpasses SOTA hard-label attacks in both attack success rate and query efficiency, particularly in low-query regimes. Beyond image classification, our approach generalizes effectively to corrupted data, biomedical datasets, and dense prediction tasks. Notably, it also successfully circumvents Blacklight, a SOTA stateful defense, resulting in a $0\%$ detection rate. Our code will be released publicly soon at https://github.com/csjunjun/DPAttack.git.

Visual token compression is widely adopted to improve the inference efficiency of Large Vision-Language Models (LVLMs), enabling their deployment in latency-sensitive and resource-constrained scenarios. However, existing work has mainly focused on efficiency and performance, while the security implications of visual token compression remain largely unexplored. In this work, we first reveal that visual token compression substantially degrades the robustness of LVLMs: models that are robust under uncompressed inference become highly vulnerable once compression is enabled. These vulnerabilities are state-specific; failure modes emerge only in the compressed setting and completely disappear when compression is disabled, making them particularly hidden and difficult to diagnose. By analyzing the key stages of the compression process, we identify instability in token importance ranking as the primary cause of this robustness degradation. Small and imperceptible perturbations can significantly alter token rankings, leading the compression mechanism to mistakenly discard task-critical information and ultimately causing model failure. Motivated by this observation, we propose a Compression-Aware Attack to systematically study and exploit this vulnerability. CAA directly targets the token selection mechanism and induces failures exclusively under compressed inference. We further extend this approach to more realistic black-box settings and introduce Transfer CAA, where neither the target model nor the compression configuration is accessible. We further evaluate potential defenses and find that they provide only limited protection. Extensive experiments across models, datasets, and compression methods show that visual token compression significantly undermines robustness, revealing a previously overlooked efficiency-security trade-off.

The rise of AI agent frameworks has introduced agent skills, modular packages containing instructions and executable code that dynamically extend agent capabilities. While this architecture enables powerful customization, skills execute with implicit trust and minimal vetting, creating a significant yet uncharacterized attack surface. We conduct the first large-scale empirical security analysis of this emerging ecosystem, collecting 42,447 skills from two major marketplaces and systematically analyzing 31,132 using SkillScan, a multi-stage detection framework integrating static analysis with LLM-based semantic classification. Our findings reveal pervasive security risks: 26.1% of skills contain at least one vulnerability, spanning 14 distinct patterns across four categories: prompt injection, data exfiltration, privilege escalation, and supply chain risks. Data exfiltration (13.3%) and privilege escalation (11.8%) are most prevalent, while 5.2% of skills exhibit high-severity patterns strongly suggesting malicious intent. We find that skills bundling executable scripts are 2.12x more likely to contain vulnerabilities than instruction-only skills (OR=2.12, p<0.001). Our contributions include: (1) a grounded vulnerability taxonomy derived from 8,126 vulnerable skills, (2) a validated detection methodology achieving 86.7% precision and 82.5% recall, and (3) an open dataset and detection toolkit to support future research. These results demonstrate an urgent need for capability-based permission systems and mandatory security vetting before this attack vector is further exploited.

Large language model (LLM)-based multi-agent systems (MAS) have shown strong capabilities in solving complex tasks. As MAS become increasingly autonomous in various safety-critical tasks, detecting malicious agents has become a critical security concern. Although existing graph anomaly detection (GAD)-based defenses can identify anomalous agents, they mainly rely on coarse sentence-level information and overlook fine-grained lexical cues, leading to suboptimal performance. Moreover, the lack of interpretability in these methods limits their reliability and real-world applicability. To address these limitations, we propose XG-Guard, an explainable and fine-grained safeguarding framework for detecting malicious agents in MAS. To incorporate both coarse and fine-grained textual information for anomalous agent identification, we utilize a bi-level agent encoder to jointly model the sentence- and token-level representations of each agent. A theme-based anomaly detector further captures the evolving discussion focus in MAS dialogues, while a bi-level score fusion mechanism quantifies token-level contributions for explanation. Extensive experiments across diverse MAS topologies and attack scenarios demonstrate robust detection performance and strong interpretability of XG-Guard.

Machine unlearning is a newly popularized technique for removing specific training data from a trained model, enabling it to comply with data deletion requests. While it protects the rights of users requesting unlearning, it also introduces new privacy risks. Prior works have primarily focused on the privacy of data that has been unlearned, while the risks to retained data remain largely unexplored. To address this gap, we focus on the privacy risks of retained data and, for the first time, reveal the vulnerabilities introduced by machine unlearning under the dual-view setting, where an adversary can query both the original and the unlearned models. From an information-theoretic perspective, we introduce the concept of {privacy knowledge gain} and demonstrate that the dual-view setting allows adversaries to obtain more information than querying either model alone, thereby amplifying privacy leakage. To effectively demonstrate this threat, we propose DVIA, a Dual-View Inference Attack, which extracts membership information on retained data using black-box queries to both models. DVIA eliminates the need to train an attack model and employs a lightweight likelihood ratio inference module for efficient inference. Experiments across different datasets and model architectures validate the effectiveness of DVIA and highlight the privacy risks inherent in the dual-view setting.